# Copyright 2023 Haotian Liu # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import os import shutil from transformers import ( AutoTokenizer, AutoModelForCausalLM, AutoConfig, BitsAndBytesConfig, ) import torch from llava.model import * from llava.constants import ( DEFAULT_IMAGE_PATCH_TOKEN, DEFAULT_IM_START_TOKEN, DEFAULT_IM_END_TOKEN, ) def load_pretrained_model( model_path, model_base, model_name, load_8bit=False, load_4bit=False, device_map="auto", load_bf16=False, ): kwargs = {"device_map": device_map} if load_8bit: kwargs["load_in_8bit"] = True elif load_4bit: kwargs["load_in_4bit"] = True kwargs["quantization_config"] = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True, bnb_4bit_quant_type="nf4", ) elif load_bf16: kwargs["torch_dtype"] = torch.bfloat16 else: kwargs["torch_dtype"] = torch.float16 if "llava" in model_name.lower(): # Load LLaVA model if "lora" in model_name.lower() and model_base is not None: lora_cfg_pretrained = AutoConfig.from_pretrained(model_path) tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False) print("Loading LLaVA from base model...") model = LlavaLlamaForCausalLM.from_pretrained( model_base, low_cpu_mem_usage=True, config=lora_cfg_pretrained, **kwargs ) if model.get_vision_tower() is not None and not model.get_vision_tower().is_loaded: model.get_vision_tower().load_model() # if the parameters have been ever modified during model training, # then for some reason, the layer will have the correct shape # but the weight will have a wrong shape # the code below fix this weird shape mismatch issue token_num, tokem_dim = model.lm_head.out_features, model.lm_head.in_features if model.lm_head.weight.shape[0] != token_num: model.lm_head.weight = torch.nn.Parameter( torch.empty(token_num, tokem_dim, device=model.device, dtype=model.dtype) ) model.model.embed_tokens.weight = torch.nn.Parameter( torch.empty(token_num, tokem_dim, device=model.device, dtype=model.dtype) ) # if the parameters have been ever modified during model training, # then for some reason, the layer will have the correct shape # but the weight will have a wrong shape # the code below fix this weird shape mismatch issue if model.get_vision_tower() is not None: mm_projector_in, mm_projector_out = ( model.model.mm_projector.in_features, model.model.mm_projector.out_features, ) if ( model.model.mm_projector.weight.shape[1] != mm_projector_in or model.model.mm_projector.weight.shape[0] != mm_projector_out ): model.model.mm_projector.weight = torch.nn.Parameter( torch.empty( mm_projector_out, mm_projector_in, device=model.device, dtype=model.dtype, ) ) model.model.mm_projector.bias = torch.nn.Parameter( torch.empty(mm_projector_out, device=model.device, dtype=model.dtype) ) print("Loading additional LLaVA weights...") if os.path.exists(os.path.join(model_path, "non_lora_trainables.bin")): non_lora_trainables = torch.load( os.path.join(model_path, "non_lora_trainables.bin"), map_location="cpu", ) else: # this is probably from HF Hub from huggingface_hub import hf_hub_download def load_from_hf(repo_id, filename, subfolder=None): cache_file = hf_hub_download( repo_id=repo_id, filename=filename, subfolder=subfolder ) return torch.load(cache_file, map_location="cpu") non_lora_trainables = load_from_hf(model_path, "non_lora_trainables.bin") non_lora_trainables = { (k[11:] if k.startswith("base_model.") else k): v for k, v in non_lora_trainables.items() } if any(k.startswith("model.model.") for k in non_lora_trainables): non_lora_trainables = { (k[6:] if k.startswith("model.") else k): v for k, v in non_lora_trainables.items() } model.load_state_dict(non_lora_trainables, strict=False) from peft import PeftModel print("Loading LoRA weights...") model = PeftModel.from_pretrained(model, model_path, device_map=device_map) print("Merging LoRA weights...") model = model.merge_and_unload() print("Model is loaded...") elif model_base is not None: # this may be mm projector only print("Loading LLaVA from base model...") if "mpt" in model_name.lower(): if not os.path.isfile(os.path.join(model_path, "configuration_mpt.py")): shutil.copyfile( os.path.join(model_base, "configuration_mpt.py"), os.path.join(model_path, "configuration_mpt.py"), ) tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=True) cfg_pretrained = AutoConfig.from_pretrained(model_path, trust_remote_code=True) model = LlavaMPTForCausalLM.from_pretrained( model_base, low_cpu_mem_usage=True, config=cfg_pretrained, **kwargs ) else: tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False) cfg_pretrained = AutoConfig.from_pretrained(model_path) model = LlavaLlamaForCausalLM.from_pretrained( model_base, low_cpu_mem_usage=True, config=cfg_pretrained, **kwargs ) # load mm projector weights (this include the vision tower weights too) if model.get_vision_tower() is not None: if not model.get_vision_tower().is_loaded: model.get_vision_tower().load_model() mm_projector_weights = torch.load( os.path.join(model_path, "mm_projector.bin"), map_location="cpu" ) mm_projector_weights = {k: v for k, v in mm_projector_weights.items()} model.load_state_dict( mm_projector_weights, strict=False ) # for 3d point cloud, this will load the vision tower too. else: if "mpt" in model_name.lower(): tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=True) model = LlavaMPTForCausalLM.from_pretrained( model_path, low_cpu_mem_usage=True, **kwargs ) else: tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False) model = LlavaLlamaForCausalLM.from_pretrained( model_path, low_cpu_mem_usage=True, **kwargs ) else: # Load language model if model_base is not None: # PEFT model from peft import PeftModel tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False) model = AutoModelForCausalLM.from_pretrained( model_base, torch_dtype=torch.bfloat16, low_cpu_mem_usage=True, device_map=device_map, ) print(f"Loading LoRA weights from {model_path}") model = PeftModel.from_pretrained(model, model_path, device_map=device_map) print(f"Merging weights") model = model.merge_and_unload() print("Convert to BF16...") model.to(torch.bfloat16) else: use_fast = False if "mpt" in model_name.lower(): tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=True) model = AutoModelForCausalLM.from_pretrained( model_path, low_cpu_mem_usage=True, trust_remote_code=True, **kwargs ) else: tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False) model = AutoModelForCausalLM.from_pretrained( model_path, low_cpu_mem_usage=True, **kwargs ) image_processor = None if "llava" in model_name.lower(): mm_use_im_start_end = getattr(model.config, "mm_use_im_start_end", False) mm_use_im_patch_token = getattr(model.config, "mm_use_im_patch_token", True) if mm_use_im_patch_token: tokenizer.add_tokens([DEFAULT_IMAGE_PATCH_TOKEN], special_tokens=True) if mm_use_im_start_end: tokenizer.add_tokens( [DEFAULT_IM_START_TOKEN, DEFAULT_IM_END_TOKEN], special_tokens=True ) model.resize_token_embeddings(len(tokenizer)) vision_tower = model.get_vision_tower() if vision_tower is not None: if not vision_tower.is_loaded: vision_tower.load_model() vision_tower.to(device=model.device, dtype=model.dtype) image_processor = vision_tower.image_processor if hasattr(model.config, "max_sequence_length"): context_len = model.config.max_sequence_length else: context_len = 2048 return tokenizer, model, image_processor, context_len